© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7785
    Prev
    Next
    INDEX
    Explanations

    FIG. 1 illustrates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    addle
    -0.10
    riz
    -0.10
    åŀĤ
    -0.09
    åŀ®
    -0.08
    acad
    -0.08
    anz
    -0.08
    åĽŀ头çľĭ
    -0.08
    åĽ°æĥij
    -0.08
    UEL
    -0.08
    主管
    -0.08
    POSITIVE LOGITS
    æīĢ示
    0.14
    ä¸¤å¼ł
    0.10
    âĩĵ
    0.10
    åıijå±ķåİĨç¨ĭ
    0.09
    åĩłå¼ł
    0.09
    ä»Ģä¹Īæĺ¯
    0.09
    asp
    0.09
    ä¸įçŁ¥ä¸įè§ī
    0.09
    ä¸Ģå¹ħ
    0.09
    æĢĸ
    0.09
    Activations Density 0.188%

    No Known Activations